LOCAL FOOD PRICES, SNAP PURCHASING POWER, AND CHILD … · 2018-06-27 · we provide the first estimates that leverage variation in SNAP purchasing power across markets to examine

NBER WORKING PAPER SERIES

LOCAL FOOD PRICES, SNAP PURCHASING POWER, AND CHILD HEALTH

Erin T. BronchettiGarret S. Christensen

Hilary W. Hoynes

Working Paper 24762http://www.nber.org/papers/w24762

NATIONAL BUREAU OF ECONOMIC RESEARCH1050 Massachusetts Avenue

Cambridge, MA 02138June 2018

This project was supported with a grant from the University of Kentucky Center for Poverty Research through funding by the U.S. Department of Agriculture, Economic Research Service and the Food and Nutrition Service, Agreement Number 58-5000-3-0066. The opinions and conclusions expressed herein are solely those of the author(s) and should not be construed as representing the opinions or policies of the sponsoring agencies. We thank seminar participants at NYU Wagner, Lausanne, Maryland, SOLE and the UCLA RDC Conference. Krista Ruffini provided excellent research assistance. The views expressed herein are those of the authors and do not necessarily reflect the views of the National Bureau of Economic Research.

NBER working papers are circulated for discussion and comment purposes. They have not been peer-reviewed or been subject to the review by the NBER Board of Directors that accompanies official NBER publications.

© 2018 by Erin T. Bronchetti, Garret S. Christensen, and Hilary W. Hoynes. All rights reserved. Short sections of text, not to exceed two paragraphs, may be quoted without explicit permission provided that full credit, including © notice, is given to the source.

Local Food Prices, SNAP Purchasing Power, and Child HealthErin T. Bronchetti, Garret S. Christensen, and Hilary W. HoynesNBER Working Paper No. 24762June 2018JEL No. H53,I12,I38

ABSTRACT

The Supplemental Nutrition Assistance Program (SNAP, formerly food stamps) is one of the most important elements of the social safety net. Unlike most other safety net programs, SNAP varies little across states and over time, which creates challenges for quasi-experimental evaluation. Notably, SNAP benefits are fixed across 48 states; but local food prices vary, leading to geographic variation in the real value – or purchasing power – of SNAP benefits. In this study, we provide the first estimates that leverage variation in SNAP purchasing power across markets to examine effects of SNAP on child health. We link panel data on regional food prices to National Health Interview Survey data and use a fixed effects framework to estimate the relationship between local purchasing power of SNAP and children’s health and health care utilization. We find that lower SNAP purchasing power leads to lower utilization of preventive health care and more days of school missed due to illness. We find no effect on reported health status.

Erin T. BronchettiDepartment of EconomicsSwarthmore College500 College AvenueSwarthmore, PA [email protected]

Garret S. Christensenc/o CEGA, 207 Giannini HallUniversity of CaliforniaBerkeley, CA [email protected]

Hilary W. HoynesRichard & Rhoda Goldman School of Public PolicyUniversity of California, Berkeley2607 Hearst AvenueBerkeley, CA 94720-7320and [email protected]

1

1 Introduction

The Supplemental Nutrition Assistance Program (SNAP, formerly the Food Stamp program)

is the largest food assistance program and one of the largest safety net programs in the United

States.1 SNAP plays a crucial role in reducing poverty for children in the U.S., with only the EITC

(combined with the Child Tax Credit) raising more children above poverty (Fox 2017). Eligibility

for the program is universal in that it depends only on a family’s income and assets; in 2015, 1

in 7 Americans received SNAP benefits (Ziliak 2015).

SNAP’s primary goals are to improve food security among low-income households, reduce

hunger, and increase access to a healthful diet.2 The extant literature demonstrates that the

program succeeds in reducing food insecurity among recipient households (see, e.g., Yen et al.

2008; Nord and Golla 2009; Mykerezi and Mills 2010; Ratcliffe, McKernan, and Zhang 2011;

Shaefer and Gutierrez 2011; Schmidt, Shore-Sheppard, and Watson 2016 and the recent review

by Hoynes and Schanzenbach 2016). Nonetheless, rates of food insecurity among SNAP

households remain quite high, raising the question of whether SNAP benefits are adequate to

meet the nutritional needs of recipients (Coleman-Jensen et al. 2012). Indeed, evidence

regarding how SNAP benefits impact recipients’ nutrition is more mixed (see, e.g., Yen 2010;

Gregory et al. 2013; Bronchetti, Christensen, and Hansen 2017b).

Our study provides unique and highly policy-relevant evidence on the impact of variation in

1 SNAP benefits paid in 2016 amounted to more than 66 billion dollars. The program has also grown dramatically in the years

since 1996 welfare reform, with benefits paid out almost tripling in real terms over the years in this study (1999-2010). 2 See, for example, the most recently amended authorizing legislation, the Food and Nutrition Act of 2008, available at

https://fns-prod.azureedge.net/sites/default/files/snap/Food-And-Nutrition-Act-2008.pdf.

2

the generosity of SNAP benefit levels on child health. Estimating the causal relationship

between SNAP and health is difficult because SNAP benefits and eligibility rules are legislated at

the federal level and do not vary across states, leaving few opportunities for quasi-experimental

analysis. One set of quasi-experimental studies analyzes the rollout of the food stamp program

across counties in the 1960s and 1970s and finds that food stamps leads to significant

improvements in birth outcomes (Currie and Moretti 2008; Almond, Hoynes, and Schanzenbach

2011) and access to food stamps in early childhood leads to significant improvements in adult

health (Hoynes, Schanzenbach, and Almond 2016). A second set of studies uses recent state

changes in application procedures (e.g. allowing online applications, whether there is a finger

printing requirement) as instruments for SNAP participation (Schmeiser 2012),3 though these

state policies had relatively small effects on participation (Ziliak 2015). A third approach is taken

by East (2016), who uses variation in eligibility for SNAP generated by welfare reform legislation

in the 1990s, and finds that SNAP in early childhood leads to improvements in health status at

ages 6-16. None of these studies, however, is able to shed light on how changes to legislated

SNAP benefit levels might impact health outcomes.

Our approach leverages plausibly exogenous geographic variation in the real value of SNAP

benefits to identify the effects of variation in SNAP generosity on health for a sample of children

in SNAP households. Importantly, the SNAP benefit formula is fixed across 48 states (benefits

are higher in Alaska and Hawaii) even though the price of food varies significantly across the

3 Gregory and Deb (2015) use the Medical Expenditure Panel Survey and state policy variables and find that SNAP participants

have fewer sick days and fewer doctor’s visits, but more checkup visits.

3

country (Todd et al. 2010; Todd, Leibtag, and Penberthy 2011).4 Across the continental U.S.,

maximum benefits vary only with family size; in 2018 a family of three is eligible for a maximum

benefit of $504/month regardless of the local cost of living. Though SNAP benefits are implicitly

adjusted for variation in the cost of living through allowed deductions (e.g., for housing, and

child care) in the calculation of net income, the limited available evidence indicates these

adjustments are not sufficient to equalize real benefits, particularly in high cost areas (Breen et

al. 2011). Gundersen, Kreider, and Pepper (2011) and the Institute of Medicine (2013) propose

this as an area for future research.

Higher SNAP purchasing power may impact children’s health through three possible

channels. A direct (nutrition) effect occurs if higher SNAP purchasing power leads to increases

in the quality or quantity of food. But higher SNAP purchasing power may also impact health

indirectly, facilitating households to increase consumption of other inputs into the health

production function, like health care. Finally, if additional SNAP purchasing power leads to

reductions in stress and “bandwidth poverty” (Bertrand et al 2004, Mullainathan and Shafir

2013), it may result in better compliance with activities such as getting children to school and to

the doctor for annual exams.

Linking nationally representative data from the 1999-2010 National Health Interview

Surveys (NHIS) to information on regional food prices from the Quarterly Food-at-home Price

Database (QFAHPD), we study the effect of variation in SNAP purchasing power on children’s

health care utilization and health. Our measure of SNAP purchasing power compares the

4 Studying data from the Quarterly Food at Home Price Database (QFAHPD), Todd et al. (2011) find that regional food prices

vary from 70 to 90 percent of the national average at the low end to 120 to 140 percent at the high end.

4

maximum SNAP benefit to the regional cost of the Thrifty Food Plan (TFP), a nutrition plan

constructed by the USDA to represent a nutritious diet at minimal cost and the basis for

maximum legislated SNAP benefits (i.e., maximum benefits are set to the TFP national average

cost). The QFAHPD includes information on food prices that allows us to construct an estimated

TFP price for each of 30 designated “market group” geographic areas across the U.S. We relate

child health care utilization and health outcomes to SNAP purchasing power (i.e., the ratio of

the national SNAP maximum benefit to the market group-level TFP price faced by a household)

in a fixed effects framework that controls for a number of individual-level and region

characteristics (including non-food prices in the area) and state policy variables. Identification

comes from differences across the 30 market groups in trends in the price of the TFP.

Our study contributes to the growing body of evidence on the SNAP program and its effects

in a few key ways. First, we provide new evidence on the relationship between SNAP benefit

generosity and the health and wellbeing of the SNAP population. Our findings consistently

indicate that children in market groups with lower purchasing power of SNAP utilize less

preventive/ambulatory health care. We find that a 10 percent increase in SNAP purchasing

power raises the likelihood a child has an annual checkup by 6.3 percentage points (8.1

percent) and the likelihood of any doctor’s visit by 3.1 percentage points (3.4 percent). While

lower real SNAP benefits do not result in significant declines in reported health status, we

document significant detrimental impacts on some health indicators, like the number of school

days missed due to illness, as well as on children’s food security. Summary indices corroborate

the existence of effects on health care utilization, but not health outcomes generally. We

confirm that these effects are not driven by relationships between geographic variation in food

5

prices and SNAP participation or health insurance coverage, nor are they present in placebo

samples of somewhat higher-income children and noncitizen children (who are ineligible for

SNAP).

A second contribution is methodological, in that our approach highlights a new

identification strategy for estimating effects of proposed changes in SNAP generosity on other

outcomes of interest. To our knowledge, ours is the first study to utilize variation in the real

value of SNAP as a source of identification.5 Future research could leverage geographic

variation in SNAP purchasing to examine SNAP’s impacts on nutrition, food consumption and

other spending patterns, birth outcomes, and adult health.6

We interpret our estimates as reflecting the impacts of variation in SNAP purchasing power,

rather than simply the effects of variation in local food prices.7 Variation in real SNAP

generosity may affect households differently than variation in prices to the extent that local

earnings adjust to account for higher local prices (Roback 1982; Albouy 2008; Moretti 2013),

while SNAP benefits do not. Indeed, we demonstrate that SNAP purchasing power does not

significantly impact our key health outcomes or food insecurity for samples that are ineligible

for SNAP or have very low SNAP utilization (e.g., non-citizen children, children in families with

incomes between 300 and 450 percent of the poverty line). Additionally, while our main models

include controls for regional prices of other goods (such as housing, energy, transportation,

5 In related work, Gregory and Coleman-Jensen (2013) study the direct relationship between local food prices and food insecurity for a sample of SNAP households. The authors find that SNAP participants in high-priced areas are 15-20 percent more likely to be food insecure than those in low-priced areas. 6 Bronchetti, Christensen, and Hansen (2017b) link National Household Food Acquisition and Purchase Survey (FoodAPS) data on SNAP recipients’ diets to local data on the cost of the TFP to study the effects of variation in SNAP purchasing power on nutrition among the SNAP population. 7 Throughout, our models include market area fixed effects and controls for local housing costs (HUD fair market rent) and other non-food prices.

6

etc.), our results are robust to the inclusion of these factors, suggesting that we are not simply

capturing the broader effects of living in a more or less expensive market.

More broadly, our findings point to sizeable, beneficial impacts of SNAP (and of increasing

the generosity of SNAP benefits) for children’s health care utilization, food security, and some

measures of their health, benefits which should be weighed carefully against the cost savings of

any proposed cuts to the SNAP program. These results also shed light on the expected impact

of adjusting benefit levels to account for geographic variation in food prices across market

groups. Such adjustments would likely reduce disparities in preventive/ambulatory care, school

absenteeism, and food security among low-income children, but may not lead to

contemporaneous changes in other health outcomes.

The paper proceeds as follows. The next section describes our multiple sources of data on

regional food prices, child health, food security, and SNAP participation, and Section 3 lays out

our empirical approach. Section 4 presents our main results regarding the impact of SNAP

purchasing power on children’s health care utilization and health, Section 5 explores

mechanisms and several robustness checks, and Section 6 concludes.

2 Data

In this study, we combine three sets of data to estimate the effect of SNAP on children’s

health. Below we describe the data on the price of the TFP, the National Health Interview

Survey, and the state and county control variables. Additionally, we supplement our main

analysis with administrative data on SNAP caseloads and household-level data on food

insecurity from the December Current Population Survey (CPS).

7

2.1 Regional Cost of the Thrifty Food Plan (TFP)

The Thrifty Food Plan (TFP) is a food plan constructed by the USDA, specifying foods and

amounts that represent a nutritious diet at a minimal cost. The TFP is used as the basis for

legislated maximum SNAP benefit levels. In 2016, the U.S. average weekly TFP cost was $146.90

for a family of four with two adults and two children (ages 6-8 and 9-11).8

To assign food prices to our sample of households in the NHIS, we construct data on the

regional price of the TFP using the Quarterly Food-at-Home Price Database (QFAHPD) (Todd et

al. 2010) for the years from 1999 through 2010. The QFAHPD, created by the USDA’s Economic

Research Service, uses Nielsen scanner data to compute quarterly estimates of the price of 52

food categories (e.g. three categories of fruit: fresh or frozen fruit, canned fruit, fruit juices;

nine categories of vegetables, etc.) for 35 regional market groups. The 35 market groups

covered in the QFAHPD include 26 metropolitan areas and 9 nonmetropolitan areas, though for

1999-2001 only 4 nonmetropolitan areas are captured.9 Each market group consists of a

combination of counties. We map the 52 QFAHPD food categories to the 29 TFP food categories

to create a single price estimate for the TFP for each market group and year during the full

1999-2010 period covered by the QFAHPD, following the methods in Gregory and Coleman-

8 See https://www.cnpp.usda.gov/sites/default/files/CostofFoodNov2016.pdf. (Accessed 1/28/17) 9 In 1999-2001, the QFAHPD identified one nonmetropolitan area for each of the 4 census regions (east, central, south and

west). In 2002 and later, they expanded to include nonmetropolitan areas in each of the 9 census divisions: New England, Middle Atlantic, East North Central, West North Central, South Atlantic, East South Central, West South Central, Mountain and Pacific. For comparability we use the four nonmetropolitan areas throughout. Appendix figure 1 shows these market groups.

https://www.cnpp.usda.gov/sites/default/files/CostofFoodNov2016.pdf

8

Jensen (2013).10, 11

To map the QFAHPD food group prices to the TFP food group prices in the market basket,

we use an expenditure-weighted average of the prices for the QFAHPD foods, where the

weights are the expenditure shares for the QFAHPD foods within each TFP category (most TFP

food categories consist of multiple QFAHPD food groups). We construct national expenditure

shares by averaging the shares across all market groups. To avoid confounding regional

variation in food prices with regional variation in consumption of different food categories, we

apply these national expenditure shares to each market group’s prices when constructing the

market group-level cost of the TFP.12, 13 We use the 2006 specification of the TFP, which

features food categories that are relatively closely aligned with the food categories in the

QFAHPD data (Carlson et al. 2007).

We assign each household in the NHIS to a market group-level TFP price based on the

county of residence and the year of interview. When estimating the relationship between the

10 We come very close to reproducing their estimates. As in this earlier work, we can cleanly link the QFAHPD categories to 23

of the 29 TFP categories without duplication or overlap of QFAHPD prices. The remaining six TFP categories contain foods that are accounted for in other parts of the QFAHPD TFP basket. For details on the construction of the TFP itself, see Carlson et al. (2007). 11 There are two versions of the QFAHPD: QFAHPD-1, which provides price data on 52 food groups for 1999-2006, and

QFAHPD-2, which includes prices for 54 food groups for 2004-2010. We bridge the two series by estimating the average ratio of QFAHPD-1 to QFAHPD-2 for years 2004 through 2006 for each market group. We then divide the price data for 1999-2003 (i.e. the years with information on only 52 food groups) by this ratio to put everything in consistent units. 12 We have also constructed measures of TFP cost using total national expenditure shares (as opposed to averaging the weights

across market groups) and obtain very similar estimates of the TFP and effect sizes. 13 An example (borrowed from Gregory and Coleman-Jensen (2013)) is illustrative. The TFP food category “whole fruit”

consists of two QFAHPD food groups: “fresh/frozen fruit” and “canned fruit.” In Hartford (market group 1) in the first quarter of 2002, expenditures on fresh/frozen fruit were $35.7 million, and expenditures on canned fruit were $5.8 million. This yields expenditure weights for whole fruit (in Hartford in quarter 1 2002) of 0.86 and 0.13, respectively. We then average these expenditure shares across all market groups to generate the national expenditure shares (for each item and period). In 2002, these national expenditure weights are 0.84 and 0.16 for fresh fruit and canned fruit, respectively. We apply these shares to the first-quarter 2002 prices of fresh/frozen and canned fruit in the Hartford market group ($0.218 and $0.244 per 100 grams, respectively) to compute a price for whole fruit in Hartford for the first quarter of 2002 (0.84×$0.218+0.16×$0.244 = $0.222 per 100 grams).

9

real value of SNAP benefits and health, we measure the purchasing power of SNAP using the

ratio of the maximum SNAP benefit to the TFP price faced by the household. Our main

regression models use the natural log of this ratio as the key independent variable for ease of

interpretation; however, results are qualitatively very similar when the level of the ratio is

employed instead.14

Figure 1 illustrates the variation across regions and over time in the real value of SNAP,

equal to the maximum SNAP benefit for a family of 4 divided by the regional cost of the TFP.15

Panel A displays the value of this ratio in 1999, Panel B shows its value in 2008, and Panel C

shows its value in 2010. In each case, a darker shading represents a higher SNAP/TFP ratio, or

greater SNAP purchasing power. In lower-cost areas the SNAP benefit covers up to 80 percent

of the cost of the TFP, while in higher cost areas (e.g., the west and northeast) this ratio falls to

less than 65 percent. Note that since the statutory TFP is constructed using a national average,

some areas are, by definition, likely to have SNAP benefits that exceed the cost of the TFP.

However, our purchasing power measure (maximum SNAP benefit/price of TFP) is less than 1

for all market groups. One reason for this is that the regional TFP prices from the QFAHPD are

based on average prices paid for each food category by all consumers, whereas the statutory

TFP price is based on prices paid by low-income persons. If low-income households are

shopping at different stores, or buying on sale or buying cheaper (e.g., store) brands, then the

TFP price we estimate using the QFAHPD will consistently be too high.

Our identification strategy relies not on the exact level of the TFP price (or SNAP purchasing

14 These results are available upon request. 15 An interactive version of this figure is available online at http://garretchristensen.shinyapps.io/Food_Price_Maps.

http://garretchristensen.shinyapps.io/Food_Price_Maps

10

power), but on the relative generosity across markets and differences across markets in trends

in SNAP purchasing power. Figure 2 demonstrates a strong, positive correlation between the

market area price paid by low-income households and the market area all-household price, for

a variety of food categories. In Appendix Figure 2, we compare our estimated TFP price to a

lowest-cost TFP price measure, by market area, which we construct using only the lowest-cost

QFAHPD food category within each TFP category (similar to how the statutory TFP price is

calculated). Reassuringly, we find a strong correlation (0.98) between our index and this

lowest-cost alternate measure across market areas.

Figure 1 also demonstrates noticeable changes in SNAP purchasing power within regions

over the 1999 – 2010 period. The changes in 2010 reflect, in part, the effect of the stimulus

package (ARRA), which raised the maximum SNAP benefit in the second half of 2009 and

throughout 2010. Appendix Figures 3a and 3b present trends in the TFP price and SNAP

purchasing power, respectively, for each of the market group areas. Trends in food prices and

SNAP purchasing power vary significantly across areas, with SNAP purchasing power rising

sharply in all areas with the ARRA.

Given that food stamp recipients are not limited to purchasing the TFP basket, why do we

use it? Ultimately, we are looking for a standardized index across places and over time that

allows us to trace out variation in food prices that are relevant for the low-income population.

Our investigations (Figure 2, Appendix Figure 2) indicate that the different price series are very

highly correlated across place. In the end, we view the TFP as appealing because it is the index

used by USDA in setting benefits. Furthermore, by using the price of a set basket we remove

the influence of any (endogenous) changes in food choices that recipients make in the face of

11

higher prices.

2.2 National Health Interview Survey (NHIS) Data on SNAP Children

We use restricted-access micro data from the National Health Interview Survey (NHIS) for

the years 1999-2010 to examine effects on child health and health care utilization.16 The NHIS

surveys approximately 35,000 households per year. By gaining restricted-use access to this data

we can observe the county of residence for each household in the survey. This allows us to link

respondents to regional area food prices and access detailed information on children’s health

and the characteristics of their parents and households for a large and representative national

sample. From each household with children, the survey selects one child at random (the

“sample child”) and collects more extensive and detailed information on this child’s health and

health care utilization. Several of the outcomes we study are only available in these Sample

Child files, while others (e.g., parent-reported health status) are available for all NHIS

respondents in the Person-level file.

Our primary sample includes children ages 17 and under who are citizens of the United

States. We impose the citizenship restriction because the post-welfare reform era witnessed

dramatic changes to rules regarding non-citizens’ eligibility for many social safety net programs,

including SNAP. (We analyze noncitizen children as a placebo group below.) We conduct our

main analyses on the sample of children in households who report having received SNAP

benefits in at least one of the past 12 months. For the years from 1999 through 2010, there are

16 State and county identifiers are masked in the public use NHIS data. Researchers interested in accessing the restricted

geocode data should contact Peter Meyer at [email protected].

12

44,627 such children; 18,299 of them are also interviewed as Sample Children. While the

advantage of limiting our analysis to the SNAP recipients is clear (this is the group most affected

by SNAP), non-random selection into SNAP participation would call into question a causal

interpretation of our estimates. In Section 4.1, we analyze the impact of SNAP purchasing

power on SNAP participation at the county level and document no significant relationship

between the real value of SNAP benefits and the per-capita SNAP caseload. As a robustness

check in Section 5, we also test the sensitivity of our results using an alternative sample with a

high likelihood of being on SNAP—children living with low-educated, unmarried parent(s).

Families with limited resources may respond to lower SNAP purchasing power by reducing

consumption of other goods that impact health, like ambulatory or preventive health care.

Additionally, lower SNAP purchasing power could lead to increases in stress and bandwidth

poverty, resulting in lower compliance with preventive care. Our primary measures of health

care utilization are indicators for whether the child has had a check-up in the past 12 months

and whether the child has had any doctor’s visit in the past 12 months. According to guidelines

from the American Academy of Pediatrics (AAP), children should have 6-7 preventive visits

before age 1, 3 visits per year as 1-year olds, 2 visits as 2-year olds, and at least one visit per

year for ages 3 through 17. We also analyze the relationship between SNAP purchasing power

and whether (the parent reports that) a child has delayed or forgone care due to cost in the

past 12 months. Finally, we study whether the child has visited the ER in the past year; if lower

SNAP purchasing power reduces the use of preventive/ambulatory care, we might also see

increased utilization of ER care.

We also analyze the effects of SNAP purchasing power on several direct measures of child

13

health that might respond to reduced nutrition, or to reduced consumption of other inputs in

the health production function (e.g., health care). Parental respondents report the child’s

health status on a 5-point scale (excellent, very good, good, fair, and poor); we use this

measure to construct an indicator for whether the child is in excellent or very good health. As

measures of contemporaneous health, we also study whether the child was hospitalized over

the past 12 months, the number of school days missed due to illness in the past 12 months (for

the sub-sample of school aged children), and an indicator for whether the child missed 5 or

more days of school due to illness. School attendance might also respond due to the stress and

bandwidth poverty channel. In addition, we estimate the relationship between SNAP

purchasing power and two longer-term health outcomes that may be affected by reduced

nutrition or to food insecurity: an indicator for obesity based on height and weight data (for the

subsample of children ages 12-17), and whether the child has emotional problems (defined for

the universe of children ages 4 and older).

In addition, we test both of these groups of outcomes (preventive health care utilization,

health outcomes) using summary index methods as in Kling, Liebman, and Katz (2007).17

Table 1 displays summary statistics for SNAP recipient children and for the entire population

of children. As expected, SNAP children are likely to be poor, live in single-parent households

(only a third live with both parents), and are disproportionately likely to be black or Hispanic.

Because such a high fraction (72 percent) of SNAP children receive Medicaid, the rate of

17 We create summary indices by subtracting the mean and dividing by the standard deviation of each variable, then averaging across variables within items in the index. Typically, the mean and standard deviation of a control group are used, but lacking that, we use the full sample. Note that the sample in these regressions is limited to those with full data from all included measures. For the health outcomes index this implies school age children only. Anderson (2008) explains similar indices clearly, and Hoynes, Schanzenbach, and Almond (2016) use the technique when evaluating long-run impacts of SNAP.

14

uninsurance among this sample is low, at about 7 percent. Health care utilization and health

outcomes are somewhat similar for SNAP citizen children compared to the general population

of children in the U.S. Nearly one-quarter of SNAP children went without a check-up in the past

year, but 90 percent had at least some sort of doctor’s visit during that time, and more than 5

percent report having delayed or gone without care due to its cost. However, ER utilization is

high, at over 30 percent, compared to 21 percent among the entire population. In terms of

health itself, SNAP children have similar health status, but miss more school days (5, on

average, but one-third of SNAP children missed 5 or more in the past year), and more

commonly have emotional problems (46 percent of SNAP children 4 or older compared to 27

percent in the general population).

2.3 State and County Control Variables We include several variables to control for regional policies and prices that might affect

child health and be correlated with local food prices. First, we control for local labor market

conditions with the county unemployment rate. Second, we include a summary index of state-

level SNAP policies developed by (Ganong and Liebman, Forthcoming), which incorporates

measures for simplified reporting, recertification lengths, interview format (e.g. in person or

not), call centers, online applications, Supplemental Security Income Combined Application

Project, vehicle exemptions for asset requirement, and broad-based categorical eligibility.

Third, we control for other state policies including the minimum wage, state EITC, TANF

maximum benefit guarantee amounts, and Medicaid/State Children’s Health Insurance Program

(CHIP) income eligibility limits. Finally, we control for prices of other goods by including HUD’s

15

fair market rent (measured by county as the “40th percentile of gross rents for typical, non-

substandard rental units occupied by recent movers in a local housing market”18) and regional

Consumer Price Indices (CPIs) for non-food, non-housing categories (apparel, commodities,

education, medical, recreation, services, transportation and other goods and services). These

are available for 26 metro areas; for the remaining areas, the CPI is calculated within each of

the four census regions and for four county population sizes (<50,000, 50,000-1.5 million, >1.5

million).

2.4 Supplemental Data on SNAP Caseloads and Food Insecurity We investigate the relationship between SNAP purchasing power and SNAP participation in

Section 4.1, using administrative data on county-level SNAP caseloads from the U.S.

Department of Agriculture (USDA), for the years from 1999 through 2010. We match each

county-year observation to that year’s TFP price for the market group to which the county

belongs.

To further probe mechanisms whereby variation in regional food prices may impact child

health, we supplement our main analysis by studying the relationship between SNAP

purchasing power and food insecurity.19 For this analysis we use data from the December

Current Population Survey Food Security Supplement (CPS-FSS) for the years from 2001-2010.20

18 More specifically, HUD estimates FMRs for 530 metropolitan areas and 2,045 nonmetropolitan county FMR areas. 19 Food insecurity is a household-level measure of well-being, defined as being unable to obtain, or uncertain of obtaining, an

adequate quantity and quality of food due to money or resources. Very-low food insecurity is defined as food insecurity that includes disrupted or restricted dietary patterns. Prior to 2006, very-low food insecurity was labeled “food insecurity with hunger”. 20 The December food security supplement was not collected in 1999 and 2000.

16

We identify a sample of 37,277 citizen children, ages 0 to 17, who live in households that report

receiving SNAP, and link them to market group TFP prices according to location of residence.21

3 Empirical Methods

We estimate the causal impact of variation in the real value of SNAP benefits on measures

of child health and health care utilization for children in households who report receiving SNAP

benefits during the past 12 months. Throughout, our regressions take the following form:

(1) 𝑦𝑖𝑟𝑡 = 𝛼 + 𝛽 ln (𝑆𝑁𝐴𝑃𝑀𝐴𝑋𝑡

𝑇𝐹𝑃𝑟𝑡) + 𝑋𝑖𝑟𝑡𝜃 + 𝑍𝑟𝑡𝛾 + 𝛿𝑡 + 𝜆𝑟 + 휀𝑖𝑟𝑡

where 𝑦𝑖𝑟𝑡 is the health outcome of individual i who resides in region r (market group) in time t.

The key independent variable is the natural log of the ratio of maximum SNAP benefits for a

family of four (which vary by year, but is constant across regions) to the TFP price in region r in

year t. The vector Xirt contains a set of controls for the child’s characteristics, including his/her

age (and its square), race, Hispanic ethnicity, family size, indicators for the presence of the

mother (and/or father) in the household, and interactions between indicators for the mother's

(father's) presence and the mother's (father's) education, marital status, age, and citizenship.

The state policy variables described in Section 2.3 are included in Zrt, as are a set of regional

CPIs in non-food, non-housing consumption categories. All models also include a full set of fixed

21 The public-use food security supplement files reports geographic information on all states, 217 counties, 69 primary

metropolitan statistical areas, 173 metropolitan statistical areas (MSA), 40 combined statistical areas (CSA), and 278 core-based statistical areas (CBSA) during our period of analysis. In order to assign CPS observations to a market group, we first identify states that include a single market group and assign all observations in that state to the corresponding market group. Continuing with the next most general geography (CSA), we repeat this process at increasingly more detailed geographies levels to the county identifiers. After this step, we then assign observations living in a non-metropolitan area to the rural market group based on their state of residence (for states with rural areas in a single market group). We match 83.7 percent of CPS observations to a market group using this iterative process.

17

effects for the year (δt) and market group (r).22 In all models, the standard errors are corrected

for clustering at the market group level.

One important question is how much variation in the cost of food remains after controlling

for the prices of other goods. Appendix Table 3 shows an R2 of 0.82 when regressing our main

food price measure on other prices indexes, and R2 of 0.97 after adding fixed effects. Appendix

Figure 4 plots the residuals, which exhibit a fair amount of idiosyncratic variation. What causes

these residuals (and leads to our plausibly exogenous variation) is by definition difficult to

explain but could be related to local wages and demand conditions, which grocery/outlet chains

are in a market, or local supply shocks.23

Identification in this model comes from variation in trends in the price of the Thrifty Food

Plan across market groups. As we discussed in Section 2.1 (see Figure 1), there is substantial

variation across geographic areas in the purchasing power of SNAP benefits. More importantly

for our identification strategy, these regional differences change over time, with some areas

experiencing larger increases in SNAP purchasing power from 1999 to 2010, and others

experiencing smaller increases (e.g., purchasing power in some southern metropolitan areas

increased nearly 17 percent, but only about 4.5 percent in urban New York).24

22 We have also tested models with additional controls including income, parent-reported health status, and an

indicator for insurance coverage, but due to endogeneity concerns, we do not include these in our main specification. The results are generally similar, however; see Appendix Tables 1 and 2. 23 It is similarly difficult to fully explain regional prices for other commodities such as gasoline in California. See for

example Borenstein, Bushnell, and Lewis (2004) and Borenstein (2015). 24 SNAP benefits in 2010 and 6 months of 2009 include increased benefits provided through the American Recovery and

Reinvestment Act (ARRA). ARRA benefits amounted to $62, or about a 13.6 percent increase above the base 2009 levels. Changes in SNAP purchasing power ranged from a decrease of 5.8 percent in San Francisco to 4.3 percent increase in metropolitan areas in Arkansas and Oklahoma over the 1999-2008 period.

18

4 Results

4.1 SNAP Participation We begin by analyzing the effects of SNAP purchasing power on the SNAP caseload. If

variation in the real value of SNAP leads to changes in SNAP participation, then selection may

bias our estimates of the effect of SNAP purchasing power on child health.

Using data from USDA, we construct a county panel for annual SNAP caseloads covering

1999-2010. We estimate equation (1) where the dependent variable is SNAP caseloads divided

by county population. Table 2 displays the results of six different specifications of the model.

Each includes year and market group fixed effects, as well as the natural log of the ratio of

maximum SNAP benefits to the market group TFP price. In the second column we add a control

for the county unemployment rate, which is a significant determinant of SNAP caseloads (Bitler

and Hoynes 2016) and possibly correlated with regional prices. In column 3 we add controls for

state policy variables, including for SNAP, EITC, minimum wages, TANF generosity, and

Medicaid. In column 4 we add controls for regional prices, including the county HUD fair market

rent and regional CPIs for goods other than food.

When only year and market group fixed effects are included, the estimated coefficient on

SNAP purchasing power is positive and significant, consistent with the SNAP caseload per capita

rising when the purchasing power of SNAP increases. However, once we add county

unemployment rate, in column (2), the coefficient drops substantially in magnitude and is no

longer statistically different from zero. The addition of the state policy controls (column 3) and

the regional prices (column 4) does not change the coefficient significantly. In column 5, we

extend the specification by including a market group linear time trend, which leads to little

19

change in the estimated coefficient on SNAP purchasing power. From this we conclude that

there is no significant relationship between the real value of SNAP and SNAP caseloads, and

thus we interpret our main results free of concerns about selection.

4.2 SNAP Purchasing Power and Health Care Utilization The primary goal of our study is to analyze the impacts of variation in the purchasing power

of SNAP benefits on outcomes related to child health. We begin by examining evidence for

measures of health care utilization, recognizing that families facing higher food prices may

respond to the lower real value of their SNAP benefits by reducing out-of-pocket spending on

other goods, including health care.

We present the results of this analysis in Table 3. Our primary measure of health care

utilization is an indicator for whether the child has had a check-up in the past 12 months

(column 1), which is observed only for children in the Sample Child file. We also examine

indicators for whether the child has had any doctor’s visit in the past 12 months (column 2),

and whether a child has visited an ER in the past 12 months (column 3). Whether a child has

delayed or forgone care is reported in the Person file of the NHIS so is observed for all NHIS

children under age 18; we report this estimate in column 4. The model includes fixed effects for

market group, year, individual controls, and regional controls for unemployment rate, non-food

prices, and state safety net policies (similar to column 4 of Table 2).25 The key independent

variable, representing SNAP purchasing power, is ln(SNAPMAX/TFP).

Among SNAP-recipient children, we find that increased purchasing power of SNAP

25 Individual-level controls include the child's age (and its square), whether the child is black or Hispanic, the child's family size,

indicators for the presence of the mother (and/or father) in the household, and interactions between indicators for the mother's (father's) presence and the mother's (father's) education, marital status, age, and citizenship.

20

significantly raises the likelihood a child has had a checkup in the past 12 months. A ten

percent increase in the ratio (SNAPMAX/TFP) leads to a 6.3 percentage point (or 8.1 percent)

increase in the likelihood of a checkup. We also document a smaller, but significant impact of

increased SNAP purchasing power on the probability a child has had any doctor’s visit over the

past 12 months. A ten percent increase in the purchasing power of SNAP raises the likelihood

of any doctor’s visit by 3.1 percentage points, or 3.4 percent.

The results in columns 3 and 4 indicate that SNAP purchasing power has no statistically

significant effect on whether children have visited the ER in the past 12 months, or are reported

to have delayed or forgone care due to cost. However, the coefficients are negative, suggesting

a protective effect of SNAP.

4.3 SNAP Purchasing Power and Health Outcomes Table 4 presents evidence on the extent to which variation in SNAP purchasing power

affects child health outcomes. The regression specifications include the same set of controls as

in Table 3. Note that several of the outcomes are defined only for sub-samples of children,

leading to different numbers of observations across the columns of Table 4. Specifically,

obesity is measured only for children ages 12 through 17,26 emotional problems are identified

for children ages 4 and older, and the number of school days missed is recorded only for

children age 5 and older who are in school. Parent-reported health status and hospitalization in

the past 12 months are reported for all children, but the other health outcomes are only

26 The indicator for obesity is based on BMI calculations, which are affected by some outlying height and weight

measurements. We trim the top and bottom of the BMI distribution to exclude the top and bottom percentile. In addition, height and weight information was only collected for children ages 12 and older in years 2008 through 2010. We therefore limit the sample to children ages 12-17.

21

provided for children in the Sample Child file.

We find no statistically significant relationship between SNAP purchasing power on the

indicator for the child’s (parent-reported) health status being excellent or very good, nor the

likelihood of having been hospitalized in the past year. However, we document a strong

negative and robust relationship between the real value of SNAP and the number of school

days children missed due to illness. For SNAP recipient children, a ten percent increase in SNAP

purchasing power is associated with a decrease in missed school days of just over 1 day (or a 22

percent decrease relative to the mean of approximately 5 days missed).

The evidence in Table 4 indicates that SNAP purchasing power does not significantly impact

obesity or the propensity to have emotional problems, although we note that these are longer

term health problems that often develop over time and may be less likely to respond

contemporaneously to higher area food prices. It is possible that these outcomes would be

likely to respond only after a longer, cumulative period of food insecurity, poor nutrition, or

reduced health care.

4.4 Summary Index Tests To address concerns of multiple hypothesis testing, we conduct a collective test of these

health care utilization and health outcomes by constructing summary index estimates as in

Kling, Liebman, and Katz (2007). We normalize and combine the outcomes into a health care

utilization index and a health index, changing signs when necessary so that all positive

outcomes mean more health care utilization or better health outcomes. The index for health

care utilization includes the variables for checkups, any doctor visits, delay seeking health care,

and ER visit. The summary index for health outcomes includes health status, hospitalization,

22

emotional problems, and school days missed.

Results are shown in Table 5. We find that a 10% increase in SNAP purchasing power leads

to a statistically significant 0.09 standard deviation increase in health care utilization. We find

positive but not significant effects on SNAP children’s health; a 10% increase in SNAP

purchasing power leads to a statistically insignificant 0.03 standard deviation increase of the

health measure.

Broadly, we interpret our results (in Tables 3 and 5) as suggesting that children in

households facing lower SNAP purchasing power receive less preventive and ambulatory care.

Our findings for health outcomes (Tables 4 and 5) suggest that variation in the real value of

SNAP may lead to changes in school attendance but has no overall impact on children’s

contemporaneous health. A weakness of measuring health using the number of school days

missed due to illness is that it may depend on the parent’s evaluation of the child’s health;

however, parent-reported child health status, which is also a subjective measure, does not

appear to respond to variation in the real value of SNAP. On the other hand, the number of

missed school days is perhaps the only health outcome we analyze that might be expected to

respond contemporaneously to reduced nutrition or limited use of preventive/ambulatory

health care.

5 Mechanisms and Robustness Checks

5.1 Mechanisms

We outlined three possible mechanisms for effects of SNAP purchasing power on child

health care utilization and health including: direct (nutrition) effects, indirect (other goods)

23

effects, and stress/bandwidth effects.

One test for the direct channel is to examine impacts of SNAP purchasing power on food

insecurity. Children in families facing higher SNAP purchasing power may be able to consume

more (or higher quality) food which may then lead to a reduction in food insecurity. Because

the NHIS did not provide information on food security or nutritional intake in the years of data

we analyze, we turn to data from the December food security supplement to the CPS to

estimate the impact of SNAP purchasing power on food insecurity among SNAP-recipient

children.

We display these results in Table 6.27 We find that a higher real value of SNAP benefits is

associated with an improvement in children’s food security: A 10 percent increase in SNAP

purchasing power reduces the likelihood a child is food insecure by 6.4 percentage points (a

21.8 percent decrease relative to the mean). These results are qualitatively quite similar to

those in Gregory and Coleman-Jensen (2013), which used fewer years of the same data and a

different estimation strategy. The result for very low food security (column 2) is not statistically

significant; however, we note that very low food security is a rare outcome even for SNAP

children (only 4 percent of the children in our sample are very food insecure while almost 30

percent are food insecure). In particular, very low food security requires not only that

households are uncertain of obtaining an adequate quantity and quality of food due to money

or resources, but that they also restrict or disrupt food intake because of lack of resources. It is

perhaps not surprising, then, that this more extreme outcome is not significantly responsive to

27 The regression specifications include the same set of controls as in Tables 3 and 4 except because we cannot identify

counties in the CPS, we do not control for local CPI for nonfood nor the HUD fair market rent data which are measured at the county level.

24

marginal variation in area food prices.

Overall, these results are suggestive that the direct effect may be part of the mechanism for

our findings. In addition, they confirm a well-studied and robust finding that higher SNAP

generosity leads to a reduction in food insecurity (Hoynes and Schanzenbach 2016). This is

important as it provides validation for our research design.

In Table 7 we investigate whether the impacts of SNAP on health care utilization and health

could be explained by a relationship between SNAP purchasing power and health insurance

coverage. Such a relationship would be unexpected for this sample, given that SNAP recipient

children are all likely to be income-eligible for Medicaid or CHIP. Returning to our sample of

NHIS children, we estimate equation (1), where the dependent variable is now an indicator for

whether the child is uninsured. Reassuringly, for both children in the Sample Child file and all

NHIS children, we find no statistically significant effect of SNAP purchasing power on the

likelihood a child has no health insurance. Additionally, Appendix Table 4 shows no relationship

between SNAP purchasing power and child participation in other food and nutrition programs.

Two of our most robust findings – that higher SNAP purchasing power leads to reductions in

school absences and increases in compliance with well child checkups – are consistent with the

stress and bandwidth channel. We are limited in our ability to test more directly for this

channel, though the NHIS does include mental health variables for the sample adult. These

include questions related to how often the respondent felt sad, worthless, nervous, hopeless,

etc. We analyzed these variables for sample adults who were mothers of the children in our

sample and found small and statistically insignificant effects.

25

5.2 SNAP Purchasing Power versus Local Prices

A natural check of our main results is to estimate our models for health care utilization and

health outcomes on a placebo sample of children that should not be affected by SNAP. We

present results for two groups: noncitizen children (who are not eligible for SNAP) and children

living in households with incomes that would make them ineligible for SNAP. 28 Appendix Figure

5 shows SNAP participation by bins of family income to poverty; we assign the placebo income

sample for incomes between 300-450% of poverty. If our main results reflect impacts of SNAP

on children’s health, rather than simply impacts of local food prices, we would expect SNAP

purchasing power to have no significant impact on these placebo samples.

These results are presented in Table 8. Panel A presents estimates for the higher income

sample, and Panel B presents estimates for noncitizen children. Estimated coefficients for our

key outcomes (i.e., had check-up, had any doctor’s visit, and number of school days missed) are

small and statistically insignificant, as are estimated effects on most other outcomes. Two

exceptions are that for the higher income sample, we find a statistically significant effect of

SNAP purchasing power on whether a child visited the ER in the past year and on whether a

child is obese. For the sample of noncitizen children, we find only a marginally significant,

negative coefficient for whether the child has an emotional problem. Recall, however, that

none of these outcomes was found to respond significantly to SNAP purchasing power among

SNAP recipient children. These findings are also reflected in estimates on the summary indexes

(see Appendix Table 5). There are small and statistically insignificant coefficients for noncitizen

children. The estimated effect on the health care utilization index for the higher income

28 As before, this sample is limited to children ages 0 through 17 who are citizens of the United States.

26

placebo group is half the size of our main effect, and statistically significant at the 10% level.

Appendix Table 6 shows there is no impact of SNAP purchasing power on child food insecurity

in these placebo samples. The bottom row of Appendix Table 5 shows SNAP participation in

these samples, which is low (4 percent or less) for high income children and 13 to 18 percent

for non-citizens.

Finally, we also explore the sensitivity of our findings to whether we control for non-food

regional CPI price controls (such as housing, energy, transportation, et cetera). As shown in

Appendix Table 7, our results are robust to the inclusion of these factors, suggesting that we

are not simply capturing the broader effects of living in a more or less expensive market.29

5.3 Robustness Checks

Table 9 displays the results of a series of robustness checks to our main findings regarding

the impacts of SNAP purchasing power on health care utilization and health. In panel A, we re-

estimate the models including a lead term that uses the t+1 market group TFP price. This lead

specification provides a test for the validity of our fixed effects design. If we find significant

effects of future SNAP purchasing power (while controlling for current purchasing power) we

might be concerned that we are capturing the effects of some other trend in the regions. That

is, we estimate:

(2) 𝑦𝑖𝑟𝑡 = 𝛼 + 𝛽1 ln (𝑆𝑁𝐴𝑃𝑀𝐴𝑋𝑡

𝑇𝐹𝑃𝑟𝑡) + 𝛽2 ln (

𝑆𝑁𝐴𝑃𝑀𝐴𝑋𝑡+1

𝑇𝐹𝑃𝑟,𝑡+1)𝑋𝑖𝑟𝑡𝜃 + 𝑍𝑟𝑡𝛾 + 𝛿𝑡 + 𝜆𝑟 + 휀𝑖𝑟𝑡

Only in one of the ten specifications, is the lead of SNAP purchasing power significant.

Importantly, our results for the contemporaneous effect of SNAP purchasing power are largely

29 In results not shown here, we estimated models where we dropped the non-food regional CPI price controls and the state

SNAP and other policy controls, and find very similar results.

27

unchanged: The magnitudes of the estimated coefficients for “had checkup” and “school days

missed” are quite similar to those in Tables 3 and 4. One exception is that the estimated impact

of current-period SNAP purchasing power on whether a child had any doctor’s visit in the past

12 months is a third as large and is no longer statistically significant.

The second panel of Table 9 contains results from a model that includes a set of market

group linear time trends. This approach places serious demands on the data in that

identification now must come from departures in market groups’ TFP prices from their trends

(assumed to be linear). While the main estimates for health care utilization (had checkup, had

any doctor’s visit) are qualitatively similar to those in Table 3, they are smaller in magnitude

and no longer statistically significant. The estimated impact of SNAP purchasing power on

missed school days, however, remains nearly identical in magnitude and significance to that in

Table 4.

Finally, to address concerns that inclusion in our SNAP recipient sample may be endogenous

to SNAP purchasing power, we estimate the impacts of variation in SNAP purchasing power on

health care utilization and health for a high intent-to-treat population. In particular, we identify

a sample of children living with unmarried parent(s) with less than a college education.30 The

results (in panel C of Table 9 along with the index models in column 3 of Appendix Table 5)

show qualitatively similar findings. Again, the estimated impacts on the likelihood of a checkup

and on the number of missed school days are quite similar in magnitude to those for our main

sample (although the p-value on the coefficient for missed school days rises to 0.141). The

30 Even though this is a high-ITT group, observable characteristics show that it is more advantaged, on average, than the SNAP

population.

28

estimated relationship between SNAP purchasing power and having had any doctor’s visit is

smaller and no longer statistically significant. Interestingly, we document a negative effect of

increased SNAP purchasing power on ER utilization for this somewhat higher-income sample: a

10 percent increase in the ratio (SNAPMAX/TFP) reduces the likelihood of an ER visit by 4.8

percentage points.

6 Discussion and Conclusion

In this paper we provide some of the first direct evidence on how variation in the

purchasing power of SNAP benefits affects children’s health care utilization and health

outcomes. We find evidence consistent with families adjusting to lower SNAP purchasing

power by reducing utilization of preventive/ambulatory medical care. In particular, we

document that a 10 percent increase in SNAP purchasing power increases the likelihood a child

had a check-up in the past year by 8.1 percent and increases the likelihood that children had

any doctor’s visit in the past 12 months by 3.4 percent.

We do not find much evidence that lower SNAP purchasing power causes detrimental

impacts on health status, the likelihood of a hospitalization, or other measures of physical (e.g.,

obesity) and mental health (e.g., child has emotional problems). One exception is that children

with reduced SNAP purchasing power miss significantly more days of school due to illness (22

percent more, relative to a baseline mean of 5 missed days, when SNAP purchasing power is

reduced by 10 percent). We also find that lower purchasing power of SNAP benefits results in a

greater likelihood of food insecurity.

One possible explanation for our finding stronger effects on utilization than on health itself

29

is that most of the health measures we consider are more chronic and cumulative in nature

(e.g., obesity). However, we also find no evidence of a relationship between SNAP purchasing

power and caregiver-reported health status, an outcome which could be less likely to suffer

from the same problem. A second possible interpretation of our findings is that while lower

SNAP purchasing power causes reduced health care utilization among children and negatively

affects food security, neither translates into substantial detrimental impacts on children’s

health status.

We also note that our measure of variation in the price of food is constructed using 30

market groups that perhaps mask variation in urban and rural customers who are in fact paying

different prices, thus masking why certain SNAP recipients are able to buy relatively

inexpensive food and stay relatively healthy. In related work, Bronchetti and Christensen (2017)

use food prices measured at a much finer geographic level from the National Household Food

Acquisition and Purchase Survey (FoodAPS) and demonstrate that whether SNAP benefits are

sufficient to buy the TFP depends largely on whether recipients are able to identify and travel

to the lowest-cost store in their area. Relating health and other outcomes to SNAP purchasing

power using finer geographic variation may be a fruitful research area in the future.

Finally, our results speak to whether adjusting benefit levels to account for geographic

variation in food prices across market groups would help improve child health and wellbeing.

We conclude that such adjustment would reduce disparities in child healthcare utilization and

school absenteeism in low-income households, but may not lead to significant changes in

contemporaneous health status.

30

31

References

Albouy, David. 2008. “Are Big Cities Bad Places to Live? Estimating Quality of Life across

Metropolitan Areas.” Working Paper 14472. National Bureau of Economic Research.

https://doi.org/10.3386/w14472.

Almond, Douglas, Hilary W. Hoynes, and Diane Whitmore Schanzenbach. 2011. “Inside the

War on Poverty: The Impact of Food Stamps on Birth Outcomes.” Review of Economics

and Statistics 93 (2): 387–403. https://doi.org/10.1162/REST_a_00089.

Anderson, Michael L. 2008. “Multiple Inference and Gender Differences in the Effects of Early

Intervention: A Reevaluation of the Abecedarian, Perry Preschool, and Early Training

Projects.” Journal of the American Statistical Association 103 (484): 1481–95.

https://doi.org/10.1198/016214508000000841.

Bitler, Marianne, and Hilary Hoynes. 2016. “The More Things Change, the More They Stay the

Same? The Safety Net and Poverty in the Great Recession.” Journal of Labor Economics

34 (S1): S403–44. https://doi.org/10.1086/683096.

Bertrand, Marianne, Sendhil Mullainathan, and Eldar Shafir. 2004. “A Behavioral-Economics

View of Poverty,” with Marianne Bertrand and Eldar Shafir, American Economic Review

Papers and Proceedings, 94(2), 419-423.

Borenstein, Severin. 2015. “Why Are California’s Gasoline Prices So High?” Energy Institute

Blog (blog). September 28, 2015. https://energyathaas.wordpress.com/2015/09/28/why-

are-californias-gasoline-prices-so-high/.

Borenstein, Severin, James Bushnell, and Matthew Lewis. 2004. “Market Power in California’s

Gasoline Market,” May. https://escholarship.org/uc/item/7vq1m8mq.

Breen, Amanda B., Rachel Cahill, Stephanie Ettinger de Cuba, John Cook, and Mariana Chilton.

2011. “Real Cost of a Healthy Diet: 2011.” Children’s Health Watch.

http://www.childrenshealthwatch.org/publication/real-cost-of-a-healthy-diet-2011/.

Bronchetti, Erin, and Garret Christensen. 2017. “Are SNAP Benefits Adequate for Purchasing a

Healthy Diet? Evidence on Geographic Variation in Food Prices and the Purchasing

Power of SNAP.” http://www.ocf.berkeley.edu/~garret.

Bronchetti, Erin, Garret Christensen, and Benjamin Hansen. 2017. “Local Food Prices and

Nutrition in SNAP Households.” presented at the APPAM Fall Research Conference,

Chicago, IL, November 3. https://github.com/garretchristensen/FoodAPSAPPAM2017.

Carlson, Andrea, Mark Lino, Wen Yen Juan, Kenneth Hanson, and P. Peter Basiotis. 2007.

“Thrifty Food Plan, 2006.” CNPP-19. US Department of Agriculture, Center for

Nutrition Policy and Promotion.

http://www.cnpp.usda.gov/sites/default/files/usda_food_plans_cost_of_food/TFP2006Re

port.pdf.

Coleman-Jensen, Alisha, Mark Nord, Margaret Andrews, and Steven Carlson. 2012. “Household

Food Security in the United States in 2011.” United States Department of Agriculture,

Economic Research Service.

https://www.ers.usda.gov/webdocs/publications/45020/30967_err141.pdf?v=41165.

Currie, Janet, and Enrico Moretti. 2008. “Short and Long-Run Effects of the Introduction of

Food Stamps on Birth Outcomes in California.” In Making Americans Healthier: Social

and Economic Policy as Health Policy. New York: Russel Sage.

32

East, Chloe N. 2016. “The Effect of Food Stamps on Children’s Health: Evidence from

Immigrants’ Changing Eligibility.” University of Colorado Denver.

http://cneast.weebly.com/uploads/8/9/9/7/8997263/east_jmp.pdf.

Fox, Liana. 2017. “The Supplemental Poverty Measure: 2016.” P60-261, September, 32.

Ganong, Peter, and Jeffrey Liebman. Forthcoming. “The Decline, Rebound, and Further Rise in

SNAP Enrollment: Disentangling Business Cycle Fluctuations and Policy Changes.”

American Economic Journal: Economic Policy.

https://www.aeaweb.org/articles?id=10.1257/pol.20140016&&from=f.

Gregory, Christian A., and Alisha Coleman-Jensen. 2013. “Do High Food Prices Increase Food

Insecurity in the United States?” Applied Economic Perspectives and Policy 35 (4): 679–

707. https://doi.org/10.1093/aepp/ppt024.

Gregory, Christian A., and Partha Deb. 2015. “Does SNAP Improve Your Health?” Food Policy

50 (January): 11–19. https://doi.org/10.1016/j.foodpol.2014.09.010.

Gregory, Christian A., Michele Ver Ploeg, Margaret Andrews, and Alisha Coleman-Jensen.

2013. “Supplemental Nutrition Assistance Program (SNAP) Participation Leads to

Modest Changes in Diet Quality.” 147. Economic Research Report. USDA Economic

Research Service.

https://www.ers.usda.gov/webdocs/publications/err147/36939_err147.pdf.

Gundersen, Craig, Brent Kreider, and John Pepper. 2011. “The Economics of Food Insecurity in

the United States.” Applied Economic Perspectives and Policy 33 (3): 281–303.

Hoynes, Hilary, and Diane Whitmore Schanzenbach. 2016. “US Food and Nutrition Programs.”

In Economics of Means-Tested Transfer Programs in the United States, Volume I.

University Of Chicago Press.

http://www.press.uchicago.edu/ucp/books/book/chicago/E/bo23520704.html.

Hoynes, Hilary, Diane Whitmore Schanzenbach, and Douglas Almond. 2016. “Long-Run

Impacts of Childhood Access to the Safety Net.” American Economic Review 106 (4):

903–34. https://doi.org/10.1257/aer.20130375.

Kling, Jeffrey R, Jeffrey B Liebman, and Lawrence F Katz. 2007. “Experimental Analysis of

Neighborhood Effects.” Econometrica 75 (1): 83–119. https://doi.org/10.1111/j.1468-

0262.2007.00733.x.

Moretti, Enrico. 2013. “Real Wage Inequality.” American Economic Journal: Applied

Economics 5 (1): 65–103. https://doi.org/10.1257/app.5.1.65.

Mullainathan, Sendhil, and Eldar Shafir. 2013. Scarcity: why having too little means so much.

First edition. New York: Times Books, Henry Holt and Company.

Mykerezi, Elton, and Bradford Mills. 2010. “The Impact of Food Stamp Program Participation

on Household Food Insecurity.” American Journal of Agricultural Economics 92 (5):

1379–1391.

Nord, M., and M. Golla. 2009. “Does SNAP Decrease Food Insecurity? Untangling the Self-

Selection Effect. USDA.” United States Department of Agriculture, Economic Research

Service.

Ratcliffe, Caroline, Signe-Mary McKernan, and Sisi Zhang. 2011. “How Much Does the

Supplemental Nutrition Assistance Program Reduce Food Insecurity?” American Journal

of Agricultural Economics 93 (4): 1082–98. https://doi.org/10.1093/ajae/aar026.

Roback, Jennifer. 1982. “Wages, Rents, and the Quality of Life.” Journal of Political Economy

90 (6): 1257–78. https://doi.org/10.1086/261120.

33

Schmeiser, Maximilian D. 2012. “The Impact of Long-Term Participation in the Supplemental

Nutrition Assistance Program on Child Obesity.” Health Economics 21 (4): 386–404.

https://doi.org/10.1002/hec.1714.

Schmidt, Lucie, Lara Shore-Sheppard, and Tara Watson. 2016. “The Effect of Safety-Net

Programs on Food Insecurity.” Journal of Human Resources 51 (3): 589–614.

https://doi.org/10.3368/jhr.51.3.1013-5987R1.

Shaefer, H Luke, and Italo Gutierrez. 2011. “The Effects of Participation in the Supplemental

Nutrition Assistance Program on the Material Hardship of Low-Income Families with

Children.” Ann Arbor, MI: National Poverty Center, Gerald R. Ford School of Public

Policy, University of Michigan.

“Supplemental Nutrition Assistance Program: Examining the Evidence to Define Benefit

Adequacy.” 2013. Institute of Medicine and National Research Council.

http://www.iom.edu/Reports/2013/Supplemental-Nutrition-Assistance-Program-

Examining-the-Evidence-to-Define-Benefit-Adequacy.aspx.

Todd, Jessica E., Ephraim Leibtag, and Corttney Penberthy. 2011. “Geographic Differences in

the Relative Price of Healthy Foods.” United States Department of Agriculture,

Economic Research Service.

http://books.google.com/books?hl=en&lr=&id=4qm5sgn3u20C&oi=fnd&pg=PP5&dq=t

odd+qfahpd&ots=UPHWMF7gnM&sig=6wOvjI4JceZtjJ-d7S111pPJm2w.

Todd, Jessica E., Lisa Mancino, Ephraim S. Leibtag, and Christina Tripodo. 2010. “Methodology

behind the Quarterly Food-at-Home Price Database.” United States Department of

Agriculture, Economic Research Service. http://ideas.repec.org/p/ags/uerstb/97799.html.

Yen, Steven T. 2010. “The Effects of SNAP and WIC Programs on Nutrient Intakes of

Children.” Food Policy 35 (6): 576–83. https://doi.org/10.1016/j.foodpol.2010.05.010.

Yen, Steven T., Margaret Andrews, Zhuo Chen, and David B. Eastwood. 2008. “Food Stamp

Program Participation and Food Insecurity: An Instrumental Variables Approach.”

American Journal of Agricultural Economics 90 (1): 117–132.

Ziliak, James. 2015. “Temporary Assistance for Needy Families.” In SNAP Matters: How Food

Stamps Affect Health and Well-Being. Stanford University Press.

Ziliak, James P. 2015. “Why Are So Many Americans on Food Stamps?” In SNAP Matters: How

Food Stamps Affect Health and Well-Being, 18-. Stanford University Press.

34

Figure 1: Purchasing Power of SNAP by Market Group Panel A: 1999

Panel B: 2008

Panel C: 2010

Notes: Maps plot SNAPMAX/TFP for each of the 30 market groups identified consistently in the Quarterly Food at Home Price Database (QFAHPD).

35

Figure 2: Comparison of Prices Across Market Groups

Full Sample vs. Low-income Sample

Canned Fruit Packaged Whole Grains

Eggs Low-Fat Milk

Carbonated Beverages

36

SNAP All SNAP All

Citizen Children Sample Sample

Children Children Children

TFP price 203 205 Any check-up (12m) 0.77 0.74

(14.36) (14.18) (0.42) (0.44)

Max SNAP benefit 143 141 Any doctor's visit (12m) 0.90 0.88

(11.57) (10.17) (0.30) (0.32)

Income to poverty ratio 0.90 3.06 Any ER visit (12m) 0.32 0.21

(0.74) (2.24) (0.47) (0.41)

Child's age 7.50 8.5 Delay/forgo care (12m) 0.06 0.05

(5.09) (5.19) (0.23) (0.22)

Child is male 0.51 0.51

(0.50) (0.50)

Child is black 0.34 0.16 Health status exc. or v. good 0.71 0.70

(0.47) (0.36) (0.45) (0.46)

Child is Hispanic 0.26 0.19 Hospitalized overnight (12m) 0.09 0.08

(0.44) (0.40) (0.28) (0.26)

Mother is present 0.94 0.95 School days missed, illness (12m) 4.96 3.53

(0.24) (0.23) (9.36) (6.43)

Father is present 0.39 0.75 5+ school days missed (12m) 0.33 0.25

(0.49) (0.43) (0.47) (0.43)

Both parents 0.36 0.71 Obese 0.20 0.13

(0.48) (0.45) (0.40) (0.34)

Child receives Medicaid 0.72 0.21 Emotional problem 0.46 0.27

(0.45) (0.41) (0.76) (0.59)

Child has no health insurance 0.07 0.10

(0.25) (0.30)

Number of observations 44,627 296,779 18,299 139,268

Notes: Tables contains estimates of means of US citizen children in household receiving SNAP, weighted to account for complex

survey design. Standard errors in parentheses. Left panel are children from the person-file dataset (i.e. all children in household,

separately for SNAP and all households) while right panel shows only children from sample child file (i.e. one child per household.)

Summary Statistics for Children in NHIS

Table 1

Number of observations

Health Outcomes

Health Care UtilizationChild/Household Characteristics

37

Outcome = SNAP CASELOAD / POPULATION(1) (2) (3) (4) (5)

log(SNAPMax/TFPt) 0.091** 0.024 0.003 -0.004 0.010

(0.036) (0.089) (0.088) (0.079) (0.085)

log(SNAPMax/TFPt+1)

Observations 37,277 37,277 37,277 37,177 37,177

R2 0.299 0.497 0.514 0.539 0.544

Mean 0.111 0.111 0.111 0.111 0.111

Effect of a 10% increase in SNAP purchasing power 0.88% 0.23% 0.03% -0.04% 0.10%

Fixed effect for year, market group Yes Yes Yes Yes Yes

County UR No Yes Yes Yes Yes

State SNAP and other policy controls No No Yes Yes Yes

Regional price controls No No No Yes Yes

Linear time trend No No No No Yes

Table 2

Effect of SNAP Purchasing Power on Per-Capita SNAP Caseload

Notes: Data consists of county by year panel for 1999-2010. Results are weighted using county population. Standard errors in

parentheses are corrected for clustering at the market group level; *** p<0.01, ** p<0.05, * p<0.1. All regressions include fixed effects

for market group and year. Columns (2)-(6) add controls for local economic and policy variables: the county unemployment rate, an

index of state SNAP policies (Ganong and Liebman, 2015), the state minimum wage, EITC, and Medicaid/SCHIP income eligibility

limits, TANF generosity, as well as controls for HUD’s fair market rent, and regional CPIs for non-food, non-housing categories

(apparel, commodities, education, medical, recreation, services, transportation and other goods and services).

38

(1) (2)

Had a Doctor's

checkup visit

past 12m past 12m

log(SNAPMAX/TFP) 0.656*** 0.323**

(0.225) (0.147)

Mean of dep. var. 0.77 0.901

Effect of 10% increase in SNAP purchasing power 0.063 0.031

As a % of mean of dep. var. 8.1% 3.4%

N 18,169 18,108

R2 0.077 0.038Notes: Results from weighted OLS regressions. Standard errors in parentheses are corrected for clustering at the market group

level; *** p<0.01, ** p<0.05, * p<0.1. All regressions include controls for the child's age (and its square), whether the child is black or

Hispanic, the child's family size, indicators for the presence of the mother (and/or father) in the household, and interactions between

indicators for the mother's (father's) presence and the mother's (father's) education, marital status, age, and citizenship. All

regressions also include controls for local economic and policy variables: the county unemployment rate, an index of state SNAP

policies (Ganong and Liebman, 2015), the state minimum wage, EITC, and Medicaid/CHIP income eligibility limits, TANF generosity,

as well as controls for HUD’s fair market rent, and regional CPIs for non-food, non-housing categories (apparel, commodities,

education, medical, recreation, services, transportation and other goods and services). Finally, all models include year and market

group fixed effects. Outcomes in columns 1-3 are observed only for children in the Sample Child files.

Table 3

Effects of Variation in SNAP Purchasing Power on Children's Health Care Utilization

Children in Sample Child File

Sample: SNAP Recipient U.S. Citizen Children in the NHIS, 1999-2010

(3)

Any ER

visit

past 12m

-0.178

(0.215)

0.315

-0.017

-5.4%

18,217

0.046Notes: Results from weighted OLS regressions. Standard errors in parentheses are corrected for clustering at the market group









Table 3


Children in Sample Child File All Children


(4)

Delay or

forgo care

past 12m

-0.089

(0.092)

0.051

-0.009

-16.6%

44,626

0.022Notes: Results from weighted OLS regressions. Standard errors in parentheses are corrected for clustering at the market group









Table 3


All Children


log(SNAPMAX/TFP)

Mean of dep. var.

Effect of 10% increase in SNAP purch power

As a % of mean of dep. var.

N

R2

Notes: Results from weighted OLS regressions. Standard errors in parentheses are corrected for clustering at the market

group level; *** p<0.01, ** p<0.05, * p<0.1. All regressions include the same controls as in Table 3. Outcomes in columns 1-4

are observed only for children in the Sample Child files. Missed school days is defined only for children ages 5 and older who

attend school; information on obesity is consistently available for children ages 12-17, trimmed to exclude the top and bottom

percentile of the BMI distribution; and emotional problem defined for the universe of children ages 4 and older.

Table 4

Effects of Variation in SNAP Purchasing Power on Children's Health Outcomes

Children in Sample Child File


(1) (2) (3) (4) (5) (6)

School days 5 or more Obese Emotional Health status Hospitalized

missed due school days problem excellent or overnight

to illness missed very good past 12m

-11.43** -0.148 -0.24 0.055 -0.121 0.020

(5.374) (0.272) (0.374) (0.468) (0.199) (0.065)

4.955 0.332 0.199 0.464 0.700 0.075

-1.090 -0.014 -0.023 0.005 -0.012 0.002

-22.0% -4.2% -11.5% 1.1% -1.6% 2.6%

11420 11420 4471 10779 44,627 44,620

0.033 0.041 0.035 0.055 0.032 0.150

Notes: Results from weighted OLS regressions. Standard errors in parentheses are corrected for clustering at the market

group level; *** p<0.01, ** p<0.05, * p<0.1. All regressions include the same controls as in Table 3. Outcomes in columns 1-4

are observed only for children in the Sample Child files. Missed school days is defined only for children ages 5 and older who

attend school; information on obesity is consistently available for children ages 12-17, trimmed to exclude the top and bottom

percentile of the BMI distribution; and emotional problem defined for the universe of children ages 4 and older.

Table 4


Children in Sample Child File All NHIS Children 0-17


39

(1) (2)

Index for health

care utilzation

Index for health

outcomes

log(SNAPMax/TFP) 0.919*** 0.311

(0.298) (0.354)

Mean of dep. var. -0.003 0.032

Effect of 10% increase in SNAP purchasing power 0.088 0.030

N 18,023 9,514

R2 0.043 0.036

Table 5

Summary Index Estimates


Notes: Table features coefficients from mean effects estimates for health care utilization variable (checkups, any

doctor visits, delay seeking health care, and any ER visit) or for health outcome variables (school days missed,

emotional problem, health status, and any hospitalization). Variables are standard normalized and averaged so

coefficient represents standard deviation units. All observations are from the Sample Child file.

40

(1) (2)

Child is food Child is very

insecure food insecure

log(SNAPMax/TFPt) -0.670* 0.0856

(0.330) (0.107)


Effect of 10% increase in SNAP purchasing power -0.064 -0.008

As a % of mean of dep. var. -21.8% -19.5%

N 29,324 29,324

R2 0.033 0.021

Table 6

Effects of SNAP Purchasing Power on Food Insecurity

Sample: SNAP Recipient U.S. Citizen Children in the December CPS, 2001-2010

Notes: Results from weighted OLS regressions. Standard errors in parentheses are corrected for clustering at the

market group level; *** p<0.01, ** p<0.05, * p<0.1. All regressions include controls for the child's age (and its

square), whether the child is black or Hispanic, the child's family size, indicators for the presence of the mother

(and/or father) in the household, and interactions between indicators for the mother's (father's) presence and the

mother's (father's) education, marital status, age, and citizenship. All regressions also include controls for local

economic and policy variables: the state unemployment rate, an index of state SNAP policies (Ganong and

Liebman, 2015), the state minimum wage, EITC, and Medicaid/CHIP income eligibility limits, and TANF generosity.

Finally, all models include year and market group fixed effects.

41

Children in

Sample Child File

(1)

No Insurance

log(SNAPMax/TFPt) -0.136

(0.146)

Mean of dep. var. 0.068

Effect of 10% increase in SNAP purchasing power -0.013

As a % of mean of dep. var. -19.0%

N 18,259

R2 0.036Notes: Results from weighted OLS regressions. Standard errors in parentheses are corrected for clustering at

the market group level; *** p<0.01, ** p<0.05, * p<0.1. All regressions include controls for the child's age (and its




economic and policy variables: the county unemployment rate, an index of state SNAP policies (Ganong and

Liebman, 2015), the state minimum wage, EITC, and Medicaid/CHIP income eligibility limits, TANF generosity, as

well as controls for HUD’s fair market rent, and regional CPIs for non-food, non-housing categories (apparel,

commodities, education, medical, recreation, services, transportation and other) goods and services. Finally, all

models include year and market group fixed effects.

Effects of SNAP Purchasing Power on Health Insurance Coverage

Table 7


All NHIS

Children 0-17

(2)

No Insurance

-0.071

(0.136)

0.067

-0.007

-10.1%

44,540

0.033Notes: Results from weighted OLS regressions. Standard errors in parentheses are corrected for clustering at

the market group level; *** p<0.01, ** p<0.05, * p<0.1. All regressions include controls for the child's age (and its




economic and policy variables: the county unemployment rate, an index of state SNAP policies (Ganong and

Liebman, 2015), the state minimum wage, EITC, and Medicaid/CHIP income eligibility limits, TANF generosity, as

well as controls for HUD’s fair market rent, and regional CPIs for non-food, non-housing categories (apparel,

commodities, education, medical, recreation, services, transportation and other) goods and services. Finally, all

models include year and market group fixed effects.

Effects of SNAP Purchasing Power on Health Insurance Coverage

Table 7


All NHIS

42

PANEL A: HIGHER INCOME CHILDREN (300-450% POVERTY)

(1) (2)

Had Doctor's

checkup visit

log(SNAPMax/TFPt) 0.239 0.196

(0.208) (0.124)


Effect of 10% increase in SNAP PP 0.022 0.018


Table 8

Effects of SNAP Purchasing Power on Health Care Utilization and Health: Robustness Checks

Placebo Samples

Health Care Utilization


(3) (4) (1) (2) (3) (4) (5) (6)

Any ER Delay or School 5+ school Obese Emotional Health status Hosp.

visit forgo care days missed days missed problem exc or v good overnight

-0.292** -0.045 2.13 -0.083 0.425** 0.297 0.008 -0.014

(0.115) -(0.045) (3.135) (0.151) (0.158) (0.272) (0.104) (0.045)

0.175 0.030 3.359 0.241 0.114 0.239 0.886 0.054

-0.029 -0.006 0.203 -0.008 0.041 0.028 0.001 -0.001

-16% -18.6% 6.0% -3.3% 36.0% 11.7% 0.1% -2.3%

Table 8


Placebo Samples

Health Care Utilization Health

Children in Sample Child File Children in Sample Child File All Children

N 24,833 24,822

R2 0.092 0.035

PANEL B: NONCITIZEN CHILDREN

(1) (2)

Had Doctor's

checkup visit


(0.429) (0.363)




24,960 48,500 18,146 18,146 8,856 15,608 48,521 48,491

0.020 0.010 0.023 0.02 0.042 0.029 0.023 0.176

(3) (4) (1) (2) (3) (4) (5) (6)

Any ER Delay or School 5+ school Obese Emotional Health status Hosp.

visit forgo care days missed days missed problem exc or v good overnight

0.403 0.112 -0.700 -0.245 0.185 -0.630* -0.310 0.0284

(0.281) (0.193) (3.567) (0.192) (0.236) (0.314) (0.251) (0.064)

0.129 0.088 2.137 0.131 0.116 0.126 0.755 0.020

0.038 0.011 -0.067 -0.023 0.018 -0.060 -0.030 0.003

29.5% 12.5% -3.1% -17.6% 15.5% -47.6% -4.0% 15.0%

N 5,440 5,421

R2 0.161 0.124

Notes: Results from weighted OLS regressions. Standard errors in parentheses are corrected for clustering at the market group level; *** p<0.01, ** p<0.05, * p<0.1. All regressions include

controls for the child's age (and its square), whether the child is black or Hispanic, the child's family size, indicators for the presence of the mother (and/or father) in the household, and

interactions between indicators for the mother's (father's) presence and the mother's (father's) education, marital status, age, and citizenship. Insurance coverage not included as control in

columns 1 and 5. All regressions also include controls for local economic and policy variables: the county unemployment rate, an index of state SNAP policies (Ganong and Liebman,

2015), the state minimum wage, EITC, TANF generosity and Medicaid/CHIP income eligibility limits, as well as controls for HUD’s fair market rent, and regional CPIs for non-food, non-

housing categories (apparel, commodities, education, medical, recreation, services, transportation and other goods and services). Finally, all models include year and market group fixed

effects. Health care utilization outcomes, columns 1-3, and health outcomes, columns 1-4 are observed only for children in the Sample Child files. Panel A contains results for children ages

0 through 17 in families with incomes between 300 and 450 percent of the federal poverty line. Panel B displays results for children ages 0 through 17 who are not US citizens.

5,458 12,358 4,809 4,809 2,490 4,163 12,372 12,368

0.034 0.035 0.022 0.028 0.089 0.029 0.077 0.025

Notes: Results from weighted OLS regressions. Standard errors in parentheses are corrected for clustering at the market group level; *** p<0.01, ** p<0.05, * p<0.1. All regressions include

controls for the child's age (and its square), whether the child is black or Hispanic, the child's family size, indicators for the presence of the mother (and/or father) in the household, and

interactions between indicators for the mother's (father's) presence and the mother's (father's) education, marital status, age, and citizenship. Insurance coverage not included as control in

columns 1 and 5. All regressions also include controls for local economic and policy variables: the county unemployment rate, an index of state SNAP policies (Ganong and Liebman,

2015), the state minimum wage, EITC, TANF generosity and Medicaid/CHIP income eligibility limits, as well as controls for HUD’s fair market rent, and regional CPIs for non-food, non-

housing categories (apparel, commodities, education, medical, recreation, services, transportation and other goods and services). Finally, all models include year and market group fixed

effects. Health care utilization outcomes, columns 1-3, and health outcomes, columns 1-4 are observed only for children in the Sample Child files. Panel A contains results for children ages

0 through 17 in families with incomes between 300 and 450 percent of the federal poverty line. Panel B displays results for children ages 0 through 17 who are not US citizens.

43

Robustness Check

(1) (2)

Had Doctor's

checkup visit

log(SNAPMax/TFPt) 0.517* 0.111

(0.278) (0.161)

log(SNAPMax/TFPt+1) 0.194 0.260

(0.247) (0.192)




Table 9


Sample: SNAP-Recipient U.S. Citizen Children, 1999-2010

A. Include lead term using future TFP price

Chldren in Sample Child File

A. Health Care Utilization

All Children

(3) (4)

Any ER Delay or

visit forgo care

0.011 0.040

(0.308) (0.089)

-0.386 -0.185*

(0.238) (0.097)

0.312 0.054

0.001 0.004

0.3% 7.1%

Table 9



Chldren in Sample Child File Chldren in Sample Child File

A. Health Care Utilization B. Health Outcomes

(1) (2) (3) (4) (5) (6)

School 5+ school Obese Emotional Health status Hosp.

days missed days missed problem exc or v good overnight

-13.48** -0.161 -0.273 0.003 -0.065 0.043

(5.90) (0.327) (0.535) (0.774) (0.305) (0.060)

-4.756 -0.303 0.0717 0.266 -0.155 -0.076

(4.02) (0.295) (0.473) (0.792) (0.316) (0.089)

4.981 0.333 0.201 0.459 0.697 0.075

-1.284 -0.015 -0.026 0.000 -0.006 0.004

-25.8% -4.6% -13.0% 0.1% -0.9% 5.5%

Table 9



All ChildrenChldren in Sample Child File

B. Health Outcomes

B. Include market group-level linear time trends (1) (2)

Had Doctor's

checkup visit


(0.272) (0.196)




(3) (4)

Any ER Delay or

visit forgo care

0.0724 -0.0316

(0.315) (0.116)

0.315 0.051

0.007 -0.003

2.2% -5.9%

(1) (2) (3) (4) (5) (6)



-12.53* -0.018 -0.351 -0.098 -0.228 0.0775

(6.82) (0.289) (0.433) (0.671) (0.248) (0.064)

4.955 0.332 0.199 0.464 0.70 0.07

-1.194 -0.002 -0.034 -0.009 -0.02 0.01

-24.1% -0.5% -16.8% -2.0% -3.1% 9.9%

C. Alternate Sample: Children of Low-Educated, (1) (2)

Unmarried Parents Had Doctor's

checkup visit

log(SNAPMax/TFPt) 0.640** 0.100

(0.255) (0.161)




(3) (4)

Any ER Delay or

visit forgo care

-0.505** 0.013

(0.184) (0.086)

0.279 0.057

-0.048 0.001

-17.3% 2.1%

(1) (2) (3) (4) (5) (6)



-11.68 -0.011 0.210 -0.305 0.058 -0.047

(7.93) (0.194) (0.344) (0.427) (0.195) (0.053)

4.323 0.305 0.181 0.396 0.705 0.062

-1.113 -0.001 0.020 -0.029 0.006 -0.005

-25.7% -0.3% 11.0% -7.3% 0.8% -7.2%

Notes: Results from weighted OLS regressions. Standard errors in parentheses are corrected for clustering at the market group level; *** p<0.01, ** p<0.05, * p<0.1. All regressions include controls for the

child's age (and its square), whether the child is black or Hispanic, the child's family size, indicators for the presence of the mother (and/or father) in the household, and interactions between indicators for

the mother's (father's) presence and the mother's (father's) education, marital status, age, and citizenship. Insurance coverage not included as control in columns 1 and 5. All regressions also include

controls for local economic and policy variables: the county unemployment rate, an index of state SNAP policies (Ganong and Liebman, 2015), the state minimum wage, EITC, TANF generosity, and

Medicaid/CHIP income eligibility limits, as well as controls for HUD’s fair market rent, and regional CPIs for non-food, non-housing categories (apparel, commodities, education, medical, recreation,

services, transportation and other). Finally, all models include year and market group fixed effects. Outcomes in Panel A, columns 1-3, and outcomes on Panel B, columns 1-4 are observed only for

children in the Sample Child files.















1

Appendix Appendix Figure 1. Reprinted from Todd et al. (2010). (Government document not subject to copyright.)

2

Appendix Figure 2: Comparison of Main TFP Measure Lowest Cost Category Measure

3

Appendix Figure 3

Variation across Market Group Areas, 1999-2010

(a) Real TFP Cost 1999-2010

(b) SNAP Purchasing Power

Notes: Figure shows (a) the TFP in constant 2010 dollars and (b) the ratio of maximum SNAP benefits to market group TFP.

Highlighted regions are those with the largest increase over the period (North Florida, New York, San Francisco) and the

largest decrease (Chicago, Metro South: Little Rock, Metro Oklahoma).

4

Appendix Figure 4

Notes: Figure shows residuals from regression of market group TFP price on other price

measures and including market group and year fixed effects.

5

Appendix Figure 5: SNAP Participation Rates by Family Income-to-Poverty Ratio

6

(1) (2) (3)

Had a Doctor's Any ER

checkup visit visit

past 12m past 12m past 12m

log(SNAPMAX/TFP) 0.648*** 0.301** -0.230

(0.223) (0.146) (0.207)

Mean of dep. var. 0.770 0.901 0.315

Effect of 10% increase in SNAP purchasing power 0.062 0.029 -0.022

As a % of mean of dep. var. 8.1% 3.2% -7.0%

N 18,126 18,065 18,171

R2 0.087 0.052 0.063Notes : Results from weighted OLS regressions. Standard errors in parentheses are corrected for clustering at the market group

level; *** p<0.01, ** p<0.05, * p<0.1. All regressions include the same controls as in Table 3 of paper but also add controls for family

income, whether child has health insurance coverage, and child's health status (1-5). Finally, all models include year and market

group fixed effects. Outcomes in columns 1, 2, and 4 are observed only for children in the Sample Child files.

Appendix Table 1




Adding controls for income, parental insurance and heath status

(4)

Delay or

forgo care

past 12m

-0.084

(0.085)

0.051

-0.008

-15.7%

44,504

0.097Notes : Results from weighted OLS regressions. Standard errors in parentheses are corrected for clustering at the market group


income, whether child has health insurance coverage, and child's health status (1-5). Finally, all models include year and market

group fixed effects. Outcomes in columns 1, 2, and 4 are observed only for children in the Sample Child files.

Appendix Table 1



All Children


(1) (2) (3) (4) (5) (6)

School days 5 or more Obese Emotional Health status Hospitalized

missed due school days problem excellent or overnight

to illness missed very good past 12m

log(SNAPMAX/TFP) -12.16** -0.179 -0.338 -0.008 -0.140 0.021

(5.57) (0.257) (0.380) (0.436) (0.196) (0.065)

Mean of dep. var. 4.96 0.332 0.199 0.463 0.700 0.075

Effect of 10% increase in SNAP purch power -1.16 -0.017 -0.032 -0.001 -0.013 0.002

As a % of mean of dep. var. -23.4% -5.1% -16.1% -0.2% -1.9% 2.7%

N 11420 11420 4471 10779 44,627 44,620

R2 0.033 0.041 0.035 0.055 0.034 0.150Notes : Results from weighted OLS regressions. Standard errors in parentheses are corrected for clustering at the market group


income, whether child has health insurance coverage, and child's health status (1-5) when health status is not the outcome of

interest. Outcomes in columns (3)-(6) are observed only for children in the Sample Child files.. Outcomes in columns (3)-(6) are

observed only for children in the Sample Child files. Missed school days is defined only for children ages 5 and older who attend

school; information on obesity is consistently available for children ages 12-17, trimmed to exclude the top and bottom percentile of

the BMI distribution; and emotional problem defined for the universe of children ages 4 and older.

Appendix Table 2



All NHIS Children 0-17


7

(1) (2) (3)

VARIABLES TFP Price TFP Price TFP Price

CPI Apparel 0.235*** 0.079 0.107

(0.08) (0.08) (0.08)

CPI Commodities 0.026 0.968*** 0.904***

(0.20) (0.31) (0.31)

CPI Education 1.963*** -0.055 -0.075

(0.11) (0.14) (0.14)

CPI Housing 2.032*** 0.687** 0.562*

(0.15) (0.28) (0.30)

CPI Medical -0.046 -0.031 -0.026

(0.06 (0.05) (0.05)

CPI Other G\&S 0.286*** 0.053 0.061

(0.04) (0.05) (0.05)

CPI Recreation 0.418*** 0.059 0.090

(0.13) (0.11) (0.11)

CPI Services -2.205*** -0.526 -0.496

(0.22) (0.35) (0.36)

CPI Transportation 0.406*** -0.641*** -0.603***

(0.11) (0.20) (0.20)

State Unemployment -0.490

(0.36)

Fair Market Rent 0.012**

(0.00)

Observations 393 393 393

R2 0.822 0.974 0.974

Marketgroup FE No Yes Yes

Year FE No Yes Yes

Appendix Table 3

Relationship between Food (TFP) Price and Other Prices

Notes: Table shows regression of main price level measure (TFP by

marketgroup and year) on other price controls and fixed effects. Standard

errors in brackets; *** p<0.01, ** p<0.05, * p<0.1.

8

(1) (2) (3)

[Ages 0-5] [Ages 5-17] [Ages 5-17]

WIC School School

Breakfast Lunch

log(SNAPMax/TFPt) 0.199 0.295 0.360

(0.493) (0.566) (0.332)

Mean of dep. var. 0.597 0.719 0.878

Effect of 10% increase in SNAP purchasing power 0.019 0.028 0.034

As a % of mean of dep. var. 3.2% 3.9% 3.9%

N 9,713 19,171 19,244

R2 0.096 0.082 0.058

Appendix Table 4

Effects of SNAP Purchasing Power on Other Program Participation

Sample: SNAP Recipient U.S. Citizen Children in the December CPS, 2001-2010

Notes: Results from weighted OLS regressions. Standard errors in parentheses are corrected for clustering at the

market group level; *** p<0.01, ** p<0.05, * p<0.1. All regressions include controls for the child's age (and its square),

whether the child is black or Hispanic, the child's family size, indicators for the presence of the mother (and/or father) in

the household, and interactions between indicators for the mother's (father's) presence and the mother's (father's)

education, marital status, age, and citizenship. All regressions also include controls for local economic and policy

variables: the state unemployment rate, an index of state SNAP policies (Ganong and Liebman, 2015), the state

minimum wage, EITC, and Medicaid/CHIP income eligibility limits, and TANF generosity. Finally, all models include year

and market group fixed effects.

9

(1) (2) (3)

SAMPLE

Citizen Children

300-450% FPL

Noncitizen

Children

Citizen Low-

Education

Unmarried

log(SNAPMax/TFPYRM) 0.555* -0.189 0.793***

[0.272] [0.442] [0.264]

Mean of dep. var. 0.011 0.013 -0.002

Effect of 10% increase in SNAP purchasing power 0.053 -0.018 0.076

N 24,662 5,402 17,651

R2 0.046 0.132 0.049

(1) (2) (3)

SAMPLE

Citizen Children

300-450% FPL

Noncitizen

Children

Citizen Low-

Education

Unmarried

log(SNAPMax/TFPYRM) -0.189 -0.0268 0.650

(0.205) (0.436) (0.457)

Mean of dep. var. 0.046 0.020 0.033

Effect of 10% increase in SNAP purchasing power -0.018 -0.003 0.062

N 14,366 3,941 9,869

R2 0.024 0.05 0.03

Mean SNAP participation rate 0.04 0.14 0.41Notes: Table features coefficients from mean effects estimates for health care utilization variable (checkups, any

doctor visits, delay seeking health care, and any ER visit) or for health outcome variables (school days missed,

emotional problem, health status, and any hospitalization). Variables are standard normalized and averaged, so

coefficient represents standard deviation units. All observations are from the Sample Child file. The sample varies by

column.

A. Health Care Utilization

Appendix Table 5

Summary Index Estimates, Placebo and Alternative Treatment Group

Placebo Samples

Placebo Samples

B. Health Outcomes

10

Citizen Children

300-450% FPL

Non-citizen

Children

log(SNAPMax/TFPt) -0.168 0.131

(0.317) (0.112)


Effect of 10% increase in SNAP PP -0.016 0.0124

As a % of mean of dep. var. -9.0% 31.2%

Appendix Table 6

Effects of SNAP Purchasing Power on Food Insecurity

Placebo Samples

Placebo Samples

N 8,415 23,939

R2 0.093 0.038

Notes: Results from weighted OLS regressions. Standard errors in parentheses are corrected for clustering

at the market group level; *** p<0.01, ** p<0.05, * p<0.1. All regressions include controls for the child's age

(and its square), whether the child is black or Hispanic, the child's family size, indicators for the presence of

the mother (and/or father) in the household, and interactions between indicators for the mother's (father's)

presence and the mother's (father's) education, marital status, age, and citizenship. All regressions also

include controls for local economic and policy variables: the state unemployment rate, an index of state

SNAP policies (Ganong and Liebman, 2015), the state minimum wage, EITC, and Medicaid/CHIP income

eligibility limits, and TANF generosity. Panel A displays results for the sample of citizen children with family

income between 300 and 450 percent of the federal poverty line, and Panel B shows results for the sample

of non-citizen children. Finally, all models include year and market group fixed effects.

11


Placebo Samples

All Children

(1) (2) (3) (4) (1) (2) (3) (5) (6)

Had Doctor's Any ER Delay or School Obese Emotional Health status Hosp.

checkup visit visit forgo care days missed problem exc or v good overnight

log(SNAPMax/TFPt) 0.616** 0.284* -0.221 -0.071 -12.91** -0.329 -0.007 -0.127 0.024

(0.224) (0.148) (0.209) (0.087) (5.362) (0.346) (0.449) (0.197) (0.062)

Mean of dep. var. 0.770 0.901 0.315 0.051 4.956 0.199 0.463 0.700 0.075

Effect of 10% increase in SNAP PP 0.059 0.027 -0.021 -0.007 -1.230 -0.031 -0.001 -0.012 0.002

As a % of mean of dep. var. 7.6% 3.0% -6.7% -13.2% -24.8% -15.7% -0.1% -1.7% 3.1%

N 18,211 18,150 18,259 44,724 11,438 4,478 10,785 44,725 44,718

R2 0.076 0.036 0.045 0.021 0.032 0.033 0.054 0.031 0.149

N

o


Effects of Variation in SNAP Purchasing Power on Children's Health Care Utilization and Health, without CPI Controls

Appendix Table 7

Health Care Utilization Health

Children in Sample Child File Children in Sample Child File All Children

Notes: Results from weighted OLS regressions. Standard errors in parentheses are corrected for clustering at the market group level; *** p<0.01, ** p<0.05, * p<0.1. All

regressions include controls for the child's age (and its square), whether the child is black or Hispanic, the child's family size, indicators for the presence of the mother

(and/or father) in the household, and interactions between indicators for the mother's (father's) presence and the mother's (father's) education, marital status, age, and

citizenship. Regressions include controls for local economic and policy variables: the county unemployment rate, an index of state SNAP policies (Ganong and Liebman,

2015), the state minimum wage, EITC, and Medicaid/CHIP income eligibility limits, TANF generosity, but do not include controls for HUD’s fair market rent, and regional

CPIs for non-food, non-housing categories (apparel, commodities, education, medical, recreation, services, transportation and other goods and services). Finally, all models

include year and market group fixed effects. Outcomes in columns 1-3 are observed only for children in the Sample Child files.

Documents

LOCAL FOOD PRICES, SNAP PURCHASING POWER, AND CHILD … · 2018-06-27 · we provide the first estimates that leverage variation in SNAP purchasing power across markets to examine